复杂系统变电站偏磁直流分布特性及影响因素分析  

Analysis on Distribution characteristics and Influencing Factors of Substation DC Magnetic Bias in Complex System

作　　者：丁国成 陈庆涛[2] 吴兴旺 王刘芳 吴伟丽[3] 刘勇 DING Guocheng;CHEN Qingtao;WU Xingwang;WANG Liufang;WU Weili;LIU Yong(Sate Grid Electric Power Research Institute of Anhui Electric Power Company,Hefei 230000,China;State Grid,Anhui Electric Power Company,Hefei 230000,China;Xi’an University of Science&Technology,Xi’an 710054,China;Anhui ZhengGuangDian Power Technology Co.,Ltd.,Hefei 230000,China)

机构地区：[1]国网安徽省电力公司电力科学研究院,合肥230000 [2]国网安徽省电力公司,安徽合肥230000 [3]西安科技大学,西安710054 [4]安徽正广电电力技术有限公司,合肥230000

出　　处：《高压电器》2020年第8期134-142,147,共10页High Voltage Apparatus

基　　金：国家电网安徽电科院科技项目的支持;新疆自治区自然基金项目支持(2017D01C417)。

摘　　要：研究直流接地极周边交流系统变压器偏磁直流分布特征及其影响因素能够为治理变压器直流偏磁提供理论依据。首先,基于变电站偏磁直流分布模型,分析了各影响因素的作用机理;其次,利用主成分分析法对影响因素进行处理,在此基础上,提出基于神经网络的变电站偏磁直流估测模型;最后,利用某电网变电站监测直流数据对上述分析方法进行验证,分析了偏磁电流对影响因素的敏感性,结果表明,所建模型的有效性,能够为规划阶段避免或降低中性点偏磁直流提供有用参考。For the purpose of DC magnetic bias control of transformer,a substation DC magnetic bias distribution model is built to analyze the influencing mechanisms of various factors on the substation DC magnetic bias distribu⁃tion.The influencing factors are disposed by using the method of principal component analysis,and a forecast model of substation DC magnetic bias based on neural network is proposed.The present analysis method is validated using the DC monitoring data of substations in a power system,and the sensitivity of the DC magnetic bias to the influenc⁃ing factors is analyzed.The results show that the proposed model is effective,and it may serve as a useful reference to avoid or reduce the DC magnetic bias in neutral point in planning phase.

关 键 词：偏磁直流 分布特性 影响因素 预测 

分 类 号：TM63[电气工程—电力系统及自动化]